CN115099490A - Yarn quality prediction method and related device - Google Patents
Yarn quality prediction method and related device Download PDFInfo
- Publication number
- CN115099490A CN115099490A CN202210725544.9A CN202210725544A CN115099490A CN 115099490 A CN115099490 A CN 115099490A CN 202210725544 A CN202210725544 A CN 202210725544A CN 115099490 A CN115099490 A CN 115099490A
- Authority
- CN
- China
- Prior art keywords
- sru
- yarn quality
- network
- quality prediction
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 144
- 230000008569 process Effects 0.000 claims abstract description 75
- 239000013598 vector Substances 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000009987 spinning Methods 0.000 claims abstract description 25
- 239000000835 fiber Substances 0.000 claims abstract description 20
- 238000003860 storage Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 11
- 238000009960 carding Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 239000002994 raw material Substances 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 5
- 230000036632 reaction speed Effects 0.000 abstract description 5
- 239000002699 waste material Substances 0.000 abstract description 5
- 229920000742 Cotton Polymers 0.000 description 7
- 230000004048 modification Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 6
- 230000008030 elimination Effects 0.000 description 3
- 238000003379 elimination reaction Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 206010020112 Hirsutism Diseases 0.000 description 1
- 238000010923 batch production Methods 0.000 description 1
- 239000011324 bead Substances 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Primary Health Care (AREA)
- Spinning Or Twisting Of Yarns (AREA)
Abstract
The application discloses a yarn quality prediction method, which comprises the following steps: processing the feature vectors of the fibers by an SRU unit; training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence; and predicting the yarn quality through the yarn quality prediction model. The method can accurately predict the yarn quality, effectively reduce the small sample trial spinning time and raw material waste when the variety is changed, and further improve the reaction speed of spinning enterprises to market demands. The application also discloses a yarn quality prediction device, equipment and a computer readable storage medium, which have the technical effects.
Description
Technical Field
The application relates to the technical field of spinning, in particular to a yarn quality prediction method; also relates to a yarn quality prediction device, equipment and a computer readable storage medium.
Background
Spinning is a complex production and processing process with behavior variables, multiple processes and multiple links, and the final yarn quality is closely related to factors such as cotton blending schemes, production processes, equipment performance and the like. Due to the existence of factors such as diversity of raw cotton fibers, variability of process routes, experience regulation and control of equipment parameters, difference of skill levels of workers and the like, the relationship between the quality attribute of the yarn and the multi-process parameters is difficult to quantify and exhaust. Especially, the process design of the yarn in the small-batch production environment depends on the engineering experience of workers, so that the fluctuation of the product quality is high, and the stability is insufficient.
Therefore, how to accurately predict the yarn quality has become a technical problem to be solved by those skilled in the art.
Disclosure of Invention
The application aims to provide a yarn quality prediction method, which can accurately predict the yarn quality, effectively reduce the sample trial spinning time and raw material waste during variety modification, and further improve the reaction speed of spinning enterprises to market demands. Another object of the present application is to provide a yarn quality prediction device, an apparatus and a computer readable storage medium, all having the above technical effects.
In order to solve the above technical problem, the present application provides a yarn quality prediction method, including:
processing the feature vectors of the fibers by an SRU unit;
training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence;
and predicting the yarn quality through the yarn quality prediction model.
Optionally, the training of the SRU network through the output of the SRU unit and the process parameters of each preset process to obtain the yarn quality prediction model includes:
inputting the output of the SRU unit into the first-stage SRU network, and respectively inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures are in one-to-one correspondence with the SRU networks; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
Optionally, before processing the feature vector of the fiber by the SRU unit, the method further includes:
and carrying out data preprocessing on the data of the characteristic parameters in the characteristic vector.
Optionally, the method for training the SRU network according to the output of the SRU unit and the process parameters of each preset process further includes:
and carrying out data preprocessing on the data of the process parameters of the preset process.
Optionally, the data preprocessing includes:
and eliminating outliers and dirty data with missing characteristics in the data, and performing normalization processing on the remaining data.
Optionally, the preset process includes a cotton carding process, a drawing process, a roving process and a spinning process.
In order to solve the above technical problem, the present application further provides a yarn quality prediction device, including:
the characteristic vector processing module is used for processing the characteristic vector of the fiber through the SRU unit;
the SRU network training module is used for training an SRU network through the output of the SRU unit and the process parameters of each preset process to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence;
and the yarn quality prediction module is used for predicting the yarn quality through the yarn quality prediction model.
Optionally, the SRU network training module is specifically configured to:
inputting the output of the SRU unit into the first-stage SRU network, and respectively inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures are in one-to-one correspondence with the SRU networks; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
In order to solve the above technical problem, the present application further provides a yarn quality prediction apparatus, including:
a memory for storing a computer program;
a processor for implementing the steps of the yarn quality prediction method as defined in any one of the above when executing said computer program.
In order to solve the above technical problem, the present application further provides a computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the yarn quality prediction method according to any one of the above methods.
The yarn quality prediction method provided by the application comprises the following steps: processing the feature vectors of the fibers by an SRU unit; training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence; and predicting the yarn quality through the yarn quality prediction model.
Therefore, according to the yarn quality prediction method provided by the application, the time sequence data in the spinning process is subjected to deep feature mining through the SRU network expressed by multiple processes, the yarn quality prediction model is obtained through training, the yarn quality is accurately predicted through the yarn quality prediction model, the sample trial spinning time and the raw material waste during variety modification can be effectively reduced, and the reaction speed of a spinning enterprise to market demands is further improved.
The yarn quality prediction device, the yarn quality prediction equipment and the computer readable storage medium have the technical effects.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the prior art and the embodiments are briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic flow chart of a yarn quality prediction method according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of an SRU unit according to an embodiment of the present disclosure;
FIG. 3 is a schematic view of a yarn quality prediction model provided in an embodiment of the present application;
fig. 4 is a schematic diagram of an SRU network according to an embodiment of the present application;
fig. 5 is a schematic view of a yarn quality prediction device provided in an embodiment of the present application;
fig. 6 is a schematic view of a yarn quality prediction apparatus provided in an embodiment of the present application.
Detailed Description
The core of the application is to provide a yarn quality prediction method, which can accurately predict the yarn quality, effectively reduce the sample trial spinning time and raw material waste during variety modification, and further improve the reaction speed of spinning enterprises to market demands. Another core of the present application is to provide a yarn quality prediction apparatus, a device and a computer readable storage medium, all having the above technical effects.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 is a schematic flow chart of a yarn quality prediction method according to an embodiment of the present application, and referring to fig. 1, the method includes:
s101: processing the feature vectors of the fibers by an SRU unit;
the characteristic parameters of the raw cotton may include: fiber purity, micronaire value, average upper half length, uniformity, short fiber index, single fiber strength, single fiber elongation, yellowness, impurities, etc. The feature vector of the fiber can be expressed as: p ═ p (p) 1 ,p 2 ,…,p n ) T ；p 1 ……p n N characteristic parameters are represented.
Referring to fig. 2, in order to implement parallel computation, in this embodiment, an SRU (Simple round-robin Unit) Unit is used to process feature vectors of fibers, and when computing the gate state of the current time step, only the timing input of the current time step is considered, and the output dependency on the previous time step is completely eliminated, thereby improving the parallel capability.
The state equation for the SRU unit is expressed as follows:
h t ＝r t ⊙g(c t )+(1-r t )⊙x t
in the above formula, x t For time sequence input, f t To forget the door, r t To reset the gate, c t For cell state renewal, h t Is a hidden layer output, sigma (-) represents a sigmoid activation function, g (-) represents a tanh activation function, a multiplication of elements of two vectors at the same position is indicated, and x t ＝[w x ,w f ,w r ] T x t To weight the input vector, w x 、w f 、w r Are all weight vectors, b ═ 0, b f ,b r ] T As an offset vector, b f 、b r Are all bias vectors.
In some embodiments, before processing the feature vector of the fiber by the SRU unit, the processing may further include:
and performing data preprocessing on the data of the characteristic parameters in the characteristic vector.
Specifically, in order to ensure the accuracy of model training and subsequent prediction, the present embodiment first performs data preprocessing on the data before performing model training using the data.
Wherein preprocessing the data may include: and eliminating outliers and dirty data with missing characteristics in the data, and performing normalization processing on the remaining data.
Specifically, the outliers can be eliminated by using a sliding window mean or median elimination method. The outlier refers to noise collected by the sensor or data with obviously incorrect numerical value. The wild value is removed by adopting a sliding window type mean value mode: and averaging the data in the sliding window, and eliminating the data with a larger difference from the average value. For example, the window size of the sliding window is 10, so that the average value of 10 data is calculated, and the data which is more different from the average value in the 10 data is removed. The median elimination method is adopted to eliminate outliers, and the median elimination method is as follows: and taking the median of some data, and rejecting the data which is greatly different from the median. For example, the median of 10 data is taken, and data that differs greatly from the median is discarded. Dirty data with missing features is discarded. For example, dirty data that lacks timestamps, lacks data characteristics due to sensor failures, and the like. And carrying out normalization processing on the data stream which is remained after the wild value and the dirty data are removed, and generating a time series sample.
Because the industrial big data has the characteristics of fast data generation, large data volume and the like, the magnitude order of the data set cannot be influenced by rejecting outliers and discarding dirty data with missing features.
S102: training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence;
in view of the characteristics of time sequence, high spatial dimension, complex dependency relationship, multi-source isomerism and the like of industrial big data, the embodiment provides an end-to-end 2D-SRU network, namely a two-dimensional SRU network. Referring to fig. 3, the SRU network includes two dimensions of timing and depth, each dimension includes a plurality of SRU units. The output of the previous SRU unit in the same dimension is input into the next SRU unit.
In the embodiment, the internal dynamic change rule of complex data is mined by utilizing the strong feature extraction capability of the deep network structure and the time sequence memory capability of the SRU network, so that end-to-end nonlinear fitting can be realized.
It can be understood that the number of layers of the neural network increases to some extent, which may enhance the capability of the model to fit the function, but the deeper layers may cause the overfitting problem and also increase the training difficulty, making the model difficult to converge. Therefore, when the model is trained, the training cost and the model prediction effect can be comprehensively considered to determine the number of the neural network layers.
In some embodiments, the preset processes may include a carding process, a drawing process, a roving process, and a spinning process.
The technological parameters of the carding process comprise: quantitative, output speed, pressure, etc. The drawing process has the following technological parameters: quantitative, output speed, number of combinations, etc. The technological parameters of the roving process comprise: basis weight, total draft, twist, etc. The technological parameters of the spinning process comprise: count, ingot speed, total draft multiple, twist, roller gauge, bead ring type, etc.
In some embodiments, the method for training the SRU network to obtain the yarn quality prediction model by using the output of the SRU unit and the process parameters of each preset process includes:
inputting the output of the SRU unit into the first-stage SRU network, and respectively inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures are in one-to-one correspondence with the SRU networks; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
Referring to fig. 4, in a typical process: on the basis of the cotton carding process, the drawing process, the roving process and the spinning process, normalized process parameters are input according to a processing time sequence. As shown in fig. 4, along the direction indicated by the arrow, the SRU network of the first stage, the SRU network of the second stage, and the SRU network of the fourth stage, i.e., the last SRU network, of the SRU network of the third stage are arranged in sequence. The output of the SRU unit is input into the first-stage SRU network, and the technological parameters of the cotton carding process are input into the first-stage SRU network. The output of the first-stage SRU network is input into the second-stage SRU network, and the technological parameters of the drawing process are input into the second-stage SRU network. And the output of the SRU network of the second stage is input into the SRU network of the third stage, and the process parameters of the roving process are input into the SRU network of the third stage. The output of the SRU network of the third stage is input into the SRU network of the fourth stage, the process parameters of the spinning process are input into the SRU network of the fourth stage, and the SRU network of the fourth stage outputs the predicted value of a certain quality index of the yarn.
The dimension of the vector input by each process is required to be consistent. When the dimension is insufficient, 0 may be complemented.
In addition, in some embodiments, the training of the SRU network may further include, through the output of the SRU unit and the process parameters of each preset process:
and carrying out data preprocessing on the data of the process parameters of the preset process.
Preprocessing the data may include: and eliminating outliers and dirty data with missing features in the data, and performing normalization processing on the remaining data.
For the specific implementation of processing data such as removing the outliers in the data, reference may be made to the above, and this embodiment is not described herein again.
S103: and predicting the yarn quality through the yarn quality prediction model.
On the basis of obtaining the yarn quality prediction model through training, the trained yarn quality prediction model can be packaged into a nonlinear functional:
in the above formula, Q is a yarn quality index predicted by the yarn quality model under a given processing technique; p ═ p (p) 1 ,p 2 ,…,p n ) T Is an n-dimensional feature vector, in which the feature parameter p 1 ,p 2 ,…,p n Can be influenced by the factors such as raw material batch, raw material producing area, production process, equipment performance and the like; t ═ t (t) 1 ,t 2 ,t 3 ,t 4 ) T The process parameter vector t of the working procedures 1, 2, 3 and 4 ref ＝(t 1,ref ,t 2,ref ,t 3,ref ,t 4,ref ) T Is a reference process parameter vector of each procedure.
In the sample trial spinning process, for a selected group of characteristic parameters, the packaged nonlinear functional is utilized to predict the yarn quality of the spun yarn process on line. Wherein the yarn quality attributes may include: yarn strength, tenacity, elongation at break, evenness unevenness, hairiness, cotton impurities and the like.
In summary, the yarn quality prediction method provided by the present application includes: processing the feature vectors of the fibers by an SRU unit; training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence; and predicting the yarn quality through the yarn quality prediction model. Therefore, according to the yarn quality prediction method provided by the application, the time sequence data in the spinning process is subjected to deep feature mining through the SRU network expressed by multiple processes, the yarn quality prediction model is obtained through training, the yarn quality is accurately predicted through the yarn quality prediction model, the small sample spinning trial time and raw material waste during variety modification can be effectively reduced, and the reaction speed of spinning enterprises to market demands is further improved.
The present application also provides a yarn quality prediction device, which is described below and to which the method described above is mutually referred. Referring to fig. 5, fig. 5 is a schematic view of a yarn quality prediction apparatus according to an embodiment of the present application, and referring to fig. 5, the apparatus includes:
a feature vector processing module 10, configured to process feature vectors of fibers through an SRU unit;
the SRU network training module 20 is used for training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence;
and the yarn quality prediction module 30 is used for predicting the yarn quality through the yarn quality prediction model.
On the basis of the foregoing embodiment, as a specific implementation manner, the SRU network training module 20 is specifically configured to:
inputting the output of the SRU unit into the first-stage SRU network, and respectively inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures are in one-to-one correspondence with the SRU networks; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
On the basis of the above embodiment, as a specific implementation manner, the method further includes:
and the first data preprocessing module is used for preprocessing the data of the characteristic parameters in the characteristic vectors.
On the basis of the above embodiment, as a specific implementation manner, the method further includes:
and the second data preprocessing module is used for preprocessing the data of the process parameters of the preset process.
On the basis of the foregoing embodiment, as a specific implementation manner, the first data preprocessing module and the second data preprocessing module are specifically configured to:
and eliminating outliers and dirty data with missing features in the data, and performing normalization processing on the remaining data.
In addition to the above-described embodiments, as a specific implementation manner, the preset process includes a carding process, a drawing process, a roving process, and a spinning process.
The application provides a yarn quality prediction device, SRU network through the multiple operation expression carries out deep level characteristic to the chronogenesis data of spinning in-process and excavates, and the training obtains yarn quality prediction model, through the accurate prediction yarn quality of this yarn quality prediction model, the hand sample that can effectively reduce when the variety is reformed transform tries on spinning time and raw materials extravagant, and then improves the reaction rate of spinning enterprise to market demand.
The present application also provides a yarn quality prediction device, as shown with reference to fig. 6, comprising a memory 1 and a processor 2.
A memory 1 for storing a computer program;
a processor 2 for executing a computer program to implement the steps of:
processing the feature vectors of the fibers by an SRU unit; training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence; and predicting the yarn quality through the yarn quality prediction model.
For the introduction of the device provided in the present application, please refer to the above method embodiment, which is not described herein again.
The present application further provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor, performs the steps of:
processing the feature vectors of the fibers by an SRU unit; training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence; and predicting the yarn quality through the yarn quality prediction model.
The computer-readable storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
For the introduction of the computer-readable storage medium provided in the present application, please refer to the above method embodiments, which are not described herein again.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device, the apparatus and the computer-readable storage medium disclosed by the embodiments correspond to the method disclosed by the embodiments, so that the description is simple, and the relevant points can be referred to the description of the method.
Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the components and steps of the various examples have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The yarn quality prediction method, apparatus, device and computer readable storage medium provided in the present application are described in detail above. The principles and embodiments of the present application are described herein using specific examples, which are only used to help understand the method and its core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
Claims (10)
1. A method of predicting yarn quality, comprising:
processing the feature vectors of the fibers by an SRU unit;
training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units and comprises two dimensions of depth and time sequence;
and predicting the yarn quality through the yarn quality prediction model.
2. The yarn quality prediction method according to claim 1, wherein the training of the SRU network through the output of the SRU unit and the process parameters of each preset process to obtain the yarn quality prediction model comprises:
inputting the output of the SRU unit into the first-stage SRU network, and inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures correspond to the SRU networks one by one; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
3. The yarn quality prediction method of claim 1, wherein processing the feature vectors of the fibers by the SRU unit further comprises:
and performing data preprocessing on the data of the characteristic parameters in the characteristic vector.
4. The yarn quality prediction method of claim 1, wherein before training the SRU network, the output of the SRU unit and the process parameters of each preset process further comprises:
and carrying out data preprocessing on the data of the process parameters of the preset process.
5. The yarn quality prediction method according to claim 3 or 4, characterized in that the data preprocessing comprises:
and eliminating outliers and dirty data with missing features in the data, and performing normalization processing on the remaining data.
6. The method of claim 1, wherein the predetermined steps include a carding step, a drawing step, a roving step, and a spinning step.
7. A yarn quality predicting device, comprising:
the characteristic vector processing module is used for processing the characteristic vector of the fiber through the SRU unit;
the SRU network training module is used for training an SRU network through the output of the SRU unit and the process parameters of each preset procedure to obtain a yarn quality prediction model; the SRU network comprises a plurality of SRU units, and the SRU network comprises two dimensions of depth and time sequence;
and the yarn quality prediction module is used for predicting the yarn quality through the yarn quality prediction model.
8. The yarn quality prediction device of claim 7, wherein the SRU network training module is specifically configured to:
inputting the output of the SRU unit into the first-stage SRU network, and inputting the process parameters of each preset procedure into the corresponding SRU network, wherein the preset procedures correspond to the SRU networks one by one; and the output of the SRU network of the previous stage is input into the SRU network of the next stage, and the output of the SRU network of the last stage is used as a predicted value of the yarn quality.
9. A yarn quality prediction apparatus, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the yarn quality prediction method according to any one of claims 1 to 6 when executing said computer program.
10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, carries out the steps of the yarn quality prediction method according to one of the claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210725544.9A CN115099490A (en) | 2022-06-24 | 2022-06-24 | Yarn quality prediction method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210725544.9A CN115099490A (en) | 2022-06-24 | 2022-06-24 | Yarn quality prediction method and related device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115099490A true CN115099490A (en) | 2022-09-23 |
Family
ID=83292977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210725544.9A Pending CN115099490A (en) | 2022-06-24 | 2022-06-24 | Yarn quality prediction method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115099490A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116356463A (en) * | 2023-04-19 | 2023-06-30 | 吴江京奕特种纤维有限公司 | Air jet vortex spun yarn quality prediction method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169565A (en) * | 2017-04-27 | 2017-09-15 | 西安工程大学 | Yarn quality prediction method based on fireworks algorithm improvement BP neural network |
| CN112488171A (en) * | 2020-11-25 | 2021-03-12 | 江南大学 | Variable mode decomposition SRU fault prediction method |
| CN112735541A (en) * | 2021-01-15 | 2021-04-30 | 杭州中科先进技术研究院有限公司 | Sewage treatment water quality prediction method based on simple circulation unit neural network |
| CN112853560A (en) * | 2020-12-31 | 2021-05-28 | 盐城师范学院 | Global process sharing control system and method based on ring spinning yarn quality |
| CN113158572A (en) * | 2021-04-27 | 2021-07-23 | 上海交通大学 | Short-term load prediction method and device |
| CN113591374A (en) * | 2021-07-14 | 2021-11-02 | 燕山大学 | Cement product specific surface area prediction method based on convolution simple cycle unit network |
-
2022
- 2022-06-24 CN CN202210725544.9A patent/CN115099490A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169565A (en) * | 2017-04-27 | 2017-09-15 | 西安工程大学 | Yarn quality prediction method based on fireworks algorithm improvement BP neural network |
| CN112488171A (en) * | 2020-11-25 | 2021-03-12 | 江南大学 | Variable mode decomposition SRU fault prediction method |
| CN112853560A (en) * | 2020-12-31 | 2021-05-28 | 盐城师范学院 | Global process sharing control system and method based on ring spinning yarn quality |
| CN112735541A (en) * | 2021-01-15 | 2021-04-30 | 杭州中科先进技术研究院有限公司 | Sewage treatment water quality prediction method based on simple circulation unit neural network |
| CN113158572A (en) * | 2021-04-27 | 2021-07-23 | 上海交通大学 | Short-term load prediction method and device |
| CN113591374A (en) * | 2021-07-14 | 2021-11-02 | 燕山大学 | Cement product specific surface area prediction method based on convolution simple cycle unit network |
Non-Patent Citations (1)
| Title |
|---|
| 胡臻龙: "基于深度神经网络的纱线质量预测模型研究", 《中国博士学位论文全文数据库 工程科技Ⅰ辑》, pages 024 - 13 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116356463A (en) * | 2023-04-19 | 2023-06-30 | 吴江京奕特种纤维有限公司 | Air jet vortex spun yarn quality prediction method |
| CN116356463B (en) * | 2023-04-19 | 2023-12-15 | 吴江京奕特种纤维有限公司 | Air jet vortex spun yarn quality prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Makis | Multivariate Bayesian control chart | |
| US20150235133A1 (en) | Data concentration prediction device, data concentration prediction method, and recording medium recording program thereof | |
| CN112036085B (en) | Steel rolling fuel consumption parameter recommendation method, system, medium and terminal | |
| CN112699051B (en) | Variant grouping method based on relevant input variables | |
| CN115099490A (en) | Yarn quality prediction method and related device | |
| Champagnat et al. | Individual-based probabilistic models of adaptive evolution and various scaling approximations | |
| CN108205432B (en) | Real-time elimination method for observation experiment data abnormal value | |
| CN109670695B (en) | Outlier data mining-based mechanical product machining procedure abnormity parallel detection method | |
| Kandasamy et al. | Survey on chaos RNN–A root cause analysis and anomaly detection | |
| CN109887253B (en) | Correlation analysis method for petrochemical device alarm | |
| CN117237479A (en) | Product style automatic generation method, device and equipment based on diffusion model | |
| CN105468871B (en) | A kind of momenttum wheel success Envelope Analysis and application process | |
| CN113946621A (en) | Mining method of filamentation workshop data fluctuation relation based on association rule | |
| CN110515796B (en) | Cortex learning-based anomaly detection method and device and terminal equipment | |
| CN113326853A (en) | Neural network based process parameter analysis method and equipment and computer storage medium | |
| CN114138025A (en) | Method and device for controlling opening of desulfurization slurry valve and related equipment | |
| CN112148730A (en) | Method for extracting product data characteristics in batches by using matrix generalized inverse | |
| Agrawal et al. | Parallelization of industrial process control program based on the technique of differential evolution using multi-threading | |
| Chang | A design parameter-free geometric Kullback-Leibler information control chart for monitoring Bernoulli processes | |
| CN115099410B (en) | Method for excavating yarn broken end cause and related device | |
| CN118394815B (en) | Method for discovering causal structure between data based on asymmetric information flow | |
| JP7322918B2 (en) | Program, information processing device, and learning model generation method | |
| CN111266405B (en) | Plate and strip hot rolling process control method and control device | |
| CN115787160B (en) | Static electricity removal control method and system for spinning machine group | |
| CN112668167B (en) | Material parameter construction method based on small amount of experimental data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220923 |